LogForum 2022, 18 (1), 77-96

> Scientific Journal of Logistics < http://doi.org/10.17270/J.LOG.2022.645

http://www.logforum.net p-ISSN 1895-2038 e-ISSN 1734-459X

ORIGINAL PAPER

FUZZY FAILURE MODE AND EFFECT ANALYSIS MODEL FOR

OPERATIONAL SUPPLY CHAIN RISKS ASSESSMENT: AN
APPLICATION IN CANNED TUNA MANUFACTURER IN THAILAND

Detcharat Sumrit, Sirima Srisawad

Faculty of Engineering, Mahidol University, Nakhon Pathom, Thailand

ABSTRACT. Background: This study proposes a multi-criterion decision-making (MCDM) framework for operational
supply chain risks assessment based on fuzzy failure mode effect analysis model. The proposed framework attempts to overcome
some weaknesses and disadvantages of the traditional FMEA in many aspects such as (i) considering “degree of difficulty to
eliminate risks” in the assessment process, (ii) using MCDM ranking methodology instead of a risk priority number, (iii) taking
both subjective and objective weights of risk criteria into account. Application of the proposed framework used canned tuna
production in Thailand as a case study.
Methods: In this study, the operational supply chain risks assessment is treated as fuzzy MCDM problem. Subjective weights
of risk criteria are determined by experts’ judgements. Objective weights are derived by Shannon entropy method. VIKOR
approach is employed to prioritize the failure modes. A sensitivity analysis is performed to examine the robustness of the
proposed framework.
Results and conclusions: The findings from this study indicates that the most three critical FMs are “risk of product
deterioration” followed by “risk of volatility raw materials supplied” and “risk of variabilities in production processes”,
respectively. It recommends that the practitioners in canned tuna industry should give the priority to mitigate these risks.
Although the present study focuses on canned tuna industry, the other similar industries can apply this proposed framework to
assess their operational supply chain risks in the same way.

Keywords: Operational supply chain risks; FMEA; MCDM; Shannon entropy; VIKOR

INTRODUCTION efficiency of supply chain operations but also the

The rapidly changing business environment
causes manufacturers to encounter various risks
and uncertainties in managing their supply chain
activities, such as variations in the lead time of
incoming raw materials, demand volatility,
unexpected machines and equipment breakdown,
labour shortage and IT disruption [Wu et al.
2019]. These risks adversely affect not only the
desired performance outcomes of manufactures.
Organizations that seek only high performance
and neglect risk management are doomed to
failure in today’s turbulent business environment
[Fan et al. 2016]. Risks may arise from natural
disasters or man-made problems and it should
have a negative impact on industries in the form
of financial and operational difficulties that could
lead to business disruption [da Silva et al. 2020].

Copyright: Wyższa Szkoła Logistyki, Poznań, Polska

Citation: Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain
risks assessment: an application in canned tuna manufacturer in Thailand. LogForum, 18(1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
Received: 26.08.2021, Accepted: 24.09.2021, on-line: 01.03.2022
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
Principally, the sources of supply chain risks
can arise from both inside and outside of
organizations [Moktadir et al. 2018]. However,
most manufacturers have internal risk
management and often overlook significant risks
throughout their supply chains. Although some
risks are inevitable, organizations should seek
proactive mechanisms to monitor, control and
manage them to alleviate their affection
[Mohamed and Youssef 2017]. Hence, supply
chain risk management (referred as SCRM)
becomes an essential part of operations
management [Shan et al. 2020]. SCRM can be
described as the identification, assessment of risks
and development of an effective risk mitigation
plan [Butdee and Phuangsalee 2019]. Proper risk
management has a positive effect on supply chain
efficiency and it can help manufactures be more
resilient in the face of major disruptions. Supply
chain risks can be classified by the source of risk
as (i) disruption risk and (ii) operational risk.
Disruption risks come from human-made and
natural disasters such as terrorist attacks,
economic crises, earthquakes, pandemics, storms,
and floods [Nakandala et al.2017]. While,
operation risks arise from the execution of
business processes or activities in supply chain
[Heckmann et al. 2015]. Major sources of
operational risks are demand uncertainties, supply
chain volatility, market price fluctuation, and
machine and equipment breakdown [Shen et al.
2020]. These risks pose a disturbance in the
supply chain and require an appropriate
assessment to develop risk mitigation strategies
[Junaid et al. 2020]. In order to prevent the
deterioration in profitability, supply chains
management is able to accurately assess risks and
fast respond to the risk events. Hence, risk
assessment is one of the important processes in
risk management [Fan et al. 2016]. There are
many risk assessment approaches reported in the
literature. Failure Mode and Effect Analysis
(FMEA) is one of the most wildly used tools in
risk management [Panchal et al. 2018]. The aim
of the FMEA method is to proactively manage
risks against potential future risk events. The
79
basic concept of FMEA is to determine risk
priority number (RPN). There are three quantified
risks criteria as severity (S), probability of
occurrence (O), and probability of detection (D),
then multiply them as a risk priority number
(RPN) [Yazdi 2019]. Eliminating or mitigating
potential risks will be planned and implemented
based on RPN prioritization manner [Liu et al.,
2018]. The FMEA calculation procedure can be
divided into three main components: (i)
determining the critical risk threshold, (ii)
calculating the RPN, and (iii) capturing data
uncertainty [Scheu et al. 2019]. As can be seen
from various studies in literature, the risk
assessment and failure modes prioritization
procedures for FMEA can be considered as multi-
criteria decision making (MCDM) problem
[Fattahi and Khalilzadeh 2018]. In addition, fuzzy
set theory (FST) is commonly used to deal with
imprecise information in decision-making
processes [Shaker et al. 2019]. A recent example
of the application of MCDM-based FST
approaches in FMEA are summarized in Table 1.
Although FMEA technique is applied in
many real-world decision-making problems,
there has been little research on supply chain risk
assessment, especially in seafood supply chain
such as tuna industry. The tuna industry has a
complex supply chain and is highly volatile in
both demand and supply. This could increase the
operational risk of the supply chain.
Comprehensive operational supply chain risk
assessment in tuna industry has not been fully
explored from existing literature. To the best of
the author’s knowledge, no studies so far have
assessed the operational risk of the supply chain
in tuna industry. To bridge the gap, this study
proposes a new multi-criteria decision-making
framework based on FMEA model to assess
operational supply chain risk under uncertain
environment. To validate the proposed
framework, canned tuna industry in Thailand is
therefore used as a case study.
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Table 1. Recent examples of the application of MCDM-based FST approaches in FMEA

No. Authors and year of FMEA problems MCDM Methodology used

publication
1 Karatop et al. (2021)
2 Rathore et al. (2021)
3 Yener and Can (2021)
4 Nabizadeh and Khalilzadeh
(2021)
5 Pourmadadkar et al. (2020)
6 Sagnak et al. (2020)
7 Yazdi et al. (2020)
8 Yan et al. (2019)
9 Wang et al. (2019)
10 Mete (2019)
Renewable energy investment Fuzzy AHP-EDAS
Evaluation of risks in foodgrains supply chain Fuzzy VIKOR
Risk assessment of air insulated metal shielded Fuzzy AHP- MABAC
cells production
Health, safety and environment risks Fuzzy goal
assessment Programming-VIKOR
Healthcare services risk assessment and quality Fuzzy TOPSIS
enhancement
Evaluation of manufacturing equipment failure Fuzzy AHP-TODIM
in hot-dip galvanizing production process
Risk analysis on a supercritical water Fuzzy best-worst method-
gasification Data Envelopment Analysis
Risk assessment for construction of urban rail Fuzzy matter-element model
transit projects
Evaluating and prioritizing risk the potential Fuzzy Choquet integral-
failure modes of steam valve system TODIM
Assessing occupational risks in pipeline Fuzzy AHP- MOORA
construction

KOmpromisno Resenje (VIKOR) approach is

RESEARCH HIGHLIGHTS AND
CONTRIBUTIONS
This study attempts to overcome some
weaknesses and drawbacks of the traditional
FMEA method by proposing a new approach to
operational supply chain risk assessment. The
highlights of this paper and the contributions to
the literature on supply chain risk assessment and
FMEA model can be summarized as follows:
- The conventional FMEA is restricted to
using only three risk criteria as S, O and D for the
FMs rating, lacking consideration of the
difficulties in eliminating any risk exposure. In
this study, a new risk factor namely “degree of
difficulty to eliminate risks” (E) is included to
analyze FMEA.
- RPN values are normally used to measure
FMs risks level in traditional FMEA. However,
the mathematical foundation for computing RPN
(S*O*D) is controversial because it is not rational
and highly sensitive to variations in results.
Because the different combination values of S, O,
and D may produce the same RPN values, but the
different FMs can have different risk levels. In
this study, a compromise programming (CP)
approach as VIsekriterijumska optimizacija i
used instead of RPN values to assess the
operational supply chain risk. The main reasons
for using VIKOR in this study are that (i) it is able
to assess and identify gaps of FMs performance
leading to further improvements (ii) it is
straightforward and uncomplicated in
computation.
- The classical FMEA assumes risk criteria
have equal important weights in risk criteria
rating which may not be realistic in various
problems in the real-world problems. The
advantage of using VIKOR method is that the
important weight of risk criteria can be altered in
the assessment process.
- Most of the previous research used either
subjective weights (purely based on the opinion
of decision-makers) or objective weights (based
on information gathered from criteria but ignoring
opinions from decision-makers) to determine the
important weights of risk criteria. In this study,
the subjective and objective weights are
combined to make the important weights of risk
criteria more reliable.
- The proposed framework of this paper
will help the practitioners and managers in canned
tuna industry to effectively assess the operational
risks and prioritize the failure modes in supply
chain. Although this study focuses on the canned

80
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

tuna industry, the proposed framework can be where 𝜇𝑁 (𝑥) is a degree of membership within
applied to other industries in a similar procedure. the range [0,1].

The present study is organized as follows: a TRAPEZOIDAL FUZZY NUMBERS

brief of mathematical preliminaries is included in
the second part. The proposed framework for
operational supply chain risks assessment is
presented in the third part of the study. The fourth
part illustrates the application of the proposed
framework of this study. Finally, the conclusions
and future research are drawn in the fifth part.
MATHEMATICAL PRELIMINARIES
FUZZY SET THEORY
Fuzzy set theory (FST) was proposed by
Zadeh [1965] to logically map linguistic variables
to crisp variables in the decision-making
processes of human judgement. FST is wildly
used to deal with uncertain and imprecise
information in fuzzy multi-criteria decision-
making (FMCDM) problems. The main concept
of FST is defined as linguistic variables based on
a specific type of fuzzy set. A fuzzy set is
generally represented by a membership function
that assigns a degree of membership within the
range [0,1] known as fuzzy number to each
linguistic variable belonging to the fuzzy set. A
fuzzy set can be mathematically defined in terms
of fuzzy numbers as: N = {(𝑥), 𝜇𝑁 (𝑥), 𝑥 ∈ 𝑅};
(TrFN)
There are different types of fuzzy numbers
such as Triangular Fuzzy Number (TFN),
Trapezoidal Fuzzy Number (TrFN), and Gaussian
Fuzzy Number (GFN). In this study, TrFN is used
to address uncertain and imprecise information
because it is a comprehensive computation and
broadly used in various problems. A pictorial
TrFN is shown in figure 1 and the mathematical
membership function can be denoted as follows
[Wang et al. 2019]:
(𝑥 − 𝑙)/(𝑚1 − 𝑙), 𝑥 ∈ [𝑙, 𝑚1 ]
1 𝑥 ∈ [𝑚1 , 𝑚2 ]
𝜇𝑁 (𝑥) = {
(𝑟 − 𝑥)/ (𝑟 − 𝑚2 ) , 𝑥 ∈ [𝑚2 , 𝑟]
0 ; 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
(1)
*where {(𝑙, 𝑚1 , 𝑚2 , 𝑟)|𝑙, 𝑚1 , 𝑚2 , 𝑟 ∈ 𝑅; 𝑙 ≤
𝑚1 ≤ 𝑚2 ≤ 𝑟}.

𝜇𝑁 (𝑥)

𝑥
𝑥1 𝑥2 𝑥3 𝑥4

Fig. 1 Trapezoidal fuzzy number

81
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

𝑥11 𝑥12 ⋯ 𝑥1𝑛

⋯ 𝑥2𝑛
COMPUTING THE SUBJECTIVE ̃ = [ 𝑥21
𝐷
𝑥22
] (4)
WEIGHTS OF RISK ⋮ ⋮ ⋱ ⋮
𝑥𝑚1 𝑥𝑚2 ⋯ 𝑥𝑚𝑛
The fuzzy rating for subjective weight of 𝑗 𝑡ℎ
𝑠 *where 𝑥𝑖𝑗 = (𝑥𝑖𝑗1, 𝑥𝑖𝑗2, 𝑥𝑖𝑗3, 𝑥𝑖𝑗4 )
criterion is given by the expert 𝑘 𝑡ℎ be 𝑤 ̃𝑗𝑘 =
𝑠 𝑠 𝑠 𝑠
{(𝑤𝑗𝑘1 , 𝑤𝑗𝑘2 , 𝑤𝑗𝑘3 , 𝑤𝑗𝑘4 )|𝑗 = 1,2, . . , 𝑛}. Hence,
DEFUZZIFY THE AGGREGATED
the fuzzy rating for subjective weights (𝑤𝑗𝑠 ) from
FUZZY RATING MATRIX
all experts are aggregated into group as
[Shemshadi et al. 2011]: Each element in the aggregated fuzzy rating
𝑠 𝑠
matrix (𝑅̃ ) is defuzzified into crisp values as
𝑤𝑗1 = min{𝑤𝑗𝑘1 } [Shemshadi et al. 2011]:
𝑘
𝑠 1
𝑤𝑗2 = ∑𝑘𝑘=1 𝑤𝑗𝑘2
𝑠
𝑘
𝑤𝑗𝑠 = 1 (2) Defuzz(𝑥𝑖𝑗 ) =
∫ 𝜇(𝑥).𝑥𝑑𝑥
(5)
𝑠
𝑤𝑗3 = ∑𝑘𝑘=1 𝑤𝑗𝑘3
𝑠
∫ 𝜇(𝑥)𝑑𝑥
𝑘
𝑠 𝑠
{ 𝑤𝑗4 = max
𝑘
{𝑤𝑗𝑘4 }
𝑥 𝑥−𝑥𝑖𝑗1 𝑥 𝑥 𝑥𝑖𝑗4 −𝑥
∫𝑥 𝑖𝑗2(𝑥 )∙𝑥𝑑𝑥+ ∫𝑥 𝑖𝑗3 𝑥𝑑𝑥+∫𝑥 𝑖𝑗4( )∙𝑥𝑑𝑥
𝑖𝑗1 𝑖𝑗2 −𝑥𝑖𝑗1 𝑖𝑗2 𝑖𝑗3 𝑥𝑖𝑗4 −𝑥𝑖𝑗3
= 𝑥−𝑥𝑖𝑗1 𝑥𝑖𝑗4 −𝑥
*where 𝑤𝑗𝑠 = [𝑤𝑗1 𝑠 𝑠 𝑠 𝑠
, 𝑤𝑗2 , 𝑤𝑗3 , 𝑤𝑗4 ] is the 𝑥 𝑥 𝑥
∫𝑥 𝑖𝑗2 (𝑥 −𝑥 )𝑑𝑥+ ∫𝑥 𝑖𝑗3 𝑑𝑥+∫𝑥 𝑖𝑗4(𝑥 −𝑥 )𝑑𝑥
𝑖𝑗1 𝑖𝑗2 𝑖𝑗1 𝑖𝑗2 𝑖𝑗3 𝑖𝑗4 𝑖𝑗3
subjective weight of 𝑗 𝑡ℎ risk criterion.
1 2 1 2
CONSTRUCT AGGREGATED FUZZY −𝑥𝑖𝑗1 𝑥𝑖𝑗2 + 𝑥𝑖𝑗3 𝑥𝑖𝑗4 + (𝑥𝑖𝑗4 − 𝑥𝑖𝑗3 ) − (𝑥𝑖𝑗2 − 𝑥𝑖𝑗1 )
= 3 3
RATING MATRIX −𝑥𝑖𝑗1 −𝑥𝑖𝑗2 + 𝑥𝑖𝑗3 +𝑥𝑖𝑗4

The fuzzy rating score collected from expert SHANNON ENTROPY APPROACH
𝑘 𝑡ℎ for 𝑖𝑡ℎ alternative regarding 𝑗 𝑡ℎ criterion is
denoted as 𝑥̃𝑖𝑗𝑘 = (𝑥𝑖𝑗𝑘1 , 𝑥𝑖𝑗𝑘2 , 𝑥𝑖𝑗𝑘3 , 𝑥𝑖𝑗𝑘4 ). Shannon [2001] introduced the entropy
Then, the fuzzy rating scores from all experts are approach to measure the uncertainty inherited in
aggregated into a group as [Shemshadi et al. information and explain it with probability theory.
2011]: In this study, entropy is employed to determine
the objective weights of risk criteria. The entropy
𝑥𝑖𝑗1 = min{𝑥𝑖𝑗𝑘1 } procedure is presented as follows [Lee and
𝑘 Chang, 2018]:
1 𝐾
𝑥𝑖𝑗2 = ∑ 𝑥
𝐾 𝑘=1 𝑖𝑗𝑘2
𝑋̃𝑖𝑗 = 1 𝐾 Step 1: Normalize the evaluation of the
𝑥𝑖𝑗3 = ∑ 𝑥
𝐾 𝑘=1 𝑖𝑗𝑘3 decision matrix
{ 𝑥𝑖𝑗4 = max
𝑘
{𝑥𝑖𝑗𝑘4 }
(3) Based on the defuzzify aggregated fuzzy
rating matrix, all elements are normalized to
Thus, the aggregated fuzzy rating matrix render the evaluation criteria become
(𝑅̃ ) is constructed as: dimensionless as:

𝑃𝑖𝑗
𝑃𝑖𝑗 = ∑𝑗 𝑃𝑖𝑗
(6)

82
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Step 2: Calculate entropy measuring values of Step 1: Normalize the aggregated fuzzy rating
criteria matrix

The entropy measuring values (𝑒𝑗 ) of the Based on the aggregated fuzzy rating matrix
evaluation criteria are calculated as: ̃
(𝑅 ), all elements are normalized to make them
can be comparable. Then, the normalized the
𝑒𝑗 = −𝑘 ∑𝑛𝑗=1 𝑃𝑖𝑗 ln (𝑃𝑖𝑗 ) (7) aggregated fuzzy rating matrix (𝑈 = [𝑢𝑖𝑗 ]𝑚×𝑛 )
is constructed as:
*where k = (ln (𝑚))−1 and m is the
number of alternatives. 𝑥𝑖𝑗1 𝑥𝑖𝑗2 𝑥𝑖𝑗3 𝑥𝑖𝑗4
𝑢𝑖𝑗 = {( − , 𝑥 − , 𝑥 − , 𝑥 − )} for cost criterion
𝑥𝑖𝑗1 𝑖𝑗1 𝑖𝑗1 𝑖𝑗1
Step 3: Determine the divergence values (10)

The divergence values (𝑑𝑗 ) of evaluation 𝑥𝑖𝑗1 𝑥𝑖𝑗2 𝑥𝑖𝑗3 𝑥𝑖𝑗4

𝑢𝑖𝑗 = {( + , + , + , + )}for benefit
criteria are determined as: 𝑥𝑖𝑗4 𝑥𝑖𝑗4 𝑥𝑖𝑗4 𝑥𝑖𝑗4
criterion (11)
𝑑𝑗 = 1 − 𝑒𝑗 (8)
+
*where 𝑥𝑖𝑗4 = max{𝑥𝑖𝑗4 } for benefit
𝑖
Step 4: Compute the normalized weights of −
criterion, while 𝑥𝑖𝑗1 = min{𝑥𝑖𝑗1 } for cost
criteria 𝑖
criterion.
The normalized weights of evaluation
Step 2: Calculate the overall performance
criteria are computed as:
rating
𝑑𝑖𝑣𝑗
𝑤𝑗 = ∑𝑗 𝑑𝑖𝑣𝑗
(9) The overall performance ratings values of
alternatives (𝑓𝑖𝑗 ) are calculated as [Shemshadi et
FUZZY VIKOR al. 2011]:

VIKOR is one of MCDM techniques that 𝐹 = [𝑓𝑖𝑗 ]𝑚×𝑛

help decision-makers prioritize the alternatives
with respect to assessment criteria. The basic 𝑓𝑖𝑗 = 𝑑𝑒𝑓𝑢𝑧𝑧(𝑢𝑖𝑗 ⨂𝑤𝑗𝑠 ) (12)
concept of this technique is that the location of the
best alternative is close to the ideal solution. In
other words, the best one has the shortest distance
from the ideal solution. In this study, fuzzy
VIKOR is employed to prioritize the failure
modes. The computation procedure of fuzzy
VIKOR is illustrated as follows [Shemshadi et al.
2011]:

∫ 𝜇(𝑥). 𝑥𝑑𝑥
𝑓𝑖𝑗 =
∫ 𝜇(𝑥)𝑑𝑥

83
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

𝑠 𝑠
𝑠
𝑢𝑖𝑗2 𝑤𝑗2 𝑥 − 𝑤𝑗1 𝑠
𝑢𝑖𝑗3 𝑤𝑗3 𝑠
𝑢𝑖𝑗4 𝑤𝑗4 𝑢𝑖𝑗4 𝑤𝑗4 −𝑥
∫𝑢 𝑠 ( 𝑠 𝑠 ) ∙ 𝑥𝑑𝑥 + ∫𝑢 𝑤 𝑠 𝑥𝑑𝑥 + ∫𝑢 𝑤 𝑠 ( 𝑠 𝑠 ) ∙ 𝑥𝑑𝑥
𝑖𝑗1 𝑤𝑗1 𝑢𝑖𝑗2 𝑤𝑗2 − 𝑢𝑖𝑗1 𝑤𝑗1 𝑖𝑗2 𝑗2 𝑖𝑗3 𝑗3 𝑢𝑖𝑗4 𝑤𝑗4 − 𝑢𝑖𝑗3 𝑤𝑗3
𝑓𝑖𝑗 = 𝑠 𝑠
𝑠
𝑢𝑖𝑗2 𝑤𝑗2 𝑥 − 𝑢𝑖𝑗1 𝑤𝑗1 𝑠
𝑢𝑖𝑗3 𝑤𝑗3 𝑠
𝑢𝑖𝑗4 𝑤𝑗4 𝑢𝑖𝑗4 𝑤𝑗4 −𝑥
∫𝑢 𝑤 𝑠 ( 𝑠 𝑠 ) 𝑑𝑥 + ∫𝑢 𝑤 𝑠 𝑑𝑥 + ∫𝑢 𝑤 𝑠 ( 𝑠 𝑠 ) 𝑑𝑥
𝑖𝑗1 𝑗1 𝑢 𝑤
𝑖𝑗2 𝑗2 − 𝑢 𝑤
𝑖𝑗1 𝑗1 𝑖𝑗2 𝑗2 𝑖𝑗3 𝑗3 𝑢 𝑤
𝑖𝑗4 𝑗4 − 𝑢 𝑖𝑗3 𝑤𝑗3

1 2 1
𝑠 𝑠 𝑠 𝑠 𝑠 𝑠 𝑠 𝑠 2
−(𝑢𝑖𝑗1 𝑢𝑖𝑗2 )(𝑤𝑗1 𝑤𝑗2 )+(𝑢𝑖𝑗3 𝑢𝑖𝑗4 )(𝑤𝑗3 𝑤𝑗4 )+3(𝑢𝑖𝑗4 𝑤𝑗4 −𝑢𝑖𝑗3 𝑤𝑗3 ) −3(𝑢𝑖𝑗2 𝑤𝑗2 −𝑢𝑖𝑗1 𝑤𝑗1 )
𝑓𝑖𝑗 = 𝑠 −𝑢 𝑤 𝑠 +𝑢 𝑤 𝑠 +𝑢 𝑤 𝑠
−𝑢𝑖𝑗1 𝑤𝑗1 𝑖𝑗2 𝑗2 𝑖𝑗3 𝑗3 𝑖𝑗4 𝑗4

𝑠 𝑠 𝑠 𝑠
where 𝑤 𝑠 = [𝑤𝑗1 , 𝑤𝑗2 , 𝑤𝑗3 , 𝑤𝑗4 ] is the subjective weights of risk criterion.

Step 3: Determine the best ideal and the worst Step 5: Compute the values of 𝑸𝒊 for failure
ideal values modes

The best ideal value (𝑓𝑗∗ ) and the worst ideal The value of 𝑄𝑖 for each failure mode can be
values (𝑓𝑗− ) of the overall performance rating computed as:
values of alternatives (𝑓𝑖𝑗 ) can be defined as:
𝜐(𝑆 −𝑆 ∗ ) (1−𝜐)(𝑅𝑖 −𝑅∗ )
𝑄𝑖 = −𝑖 ∗ +
𝑆 −𝑆 𝑅− −𝑅∗
𝑓𝑖∗ = max{𝑓𝑖𝑗 } (13) (18)
𝑖

𝑓𝑖− = min{𝑓𝑖𝑗 } (14)

𝑖
*where
Step 4: Calculate the values of utility measure
𝑺𝒊 and regret measure 𝑹𝒊 for each alternative 𝑆 − = max{𝑆𝑖 } (19)
𝑖

The value of 𝑆𝑖 and 𝑅𝑖 for each alternative

𝑆 ∗ = min{𝑆𝑖 } (20)
can be calculated by using LP-matric (𝐿𝑃,𝑖 ) as an 𝑖
aggregating function to determine the
compromise raking of the alternative. According 𝑅 − = max{𝑅𝑖 } (21)
𝑖
to LP-matric, 𝐿1,𝑖 and 𝐿∞,𝑖 are used to compute 𝑆𝑖
and 𝑅𝑖 as follows: 𝑅 ∗ = min{𝑅𝑖 } (22)
𝑖

𝐿𝑃,𝑖 = {∑𝑛𝑗=1[𝑤𝑗 (𝑟𝑗∗ − 𝑟𝑖𝑗 )/(𝑟𝑗∗ − 𝜐 stands for weight for the strategy of
1
𝑝 ⁄𝑝 maximum group utility, while 1 − 𝜐 stands for the
𝑟𝑗− )] } (15) weight of the individual regret. In this study, the
𝜐 value is 0.5.
𝑤𝑗𝑜 (𝑓𝑖∗−𝑓𝑖𝑗 )
𝑆𝑖 = ∑𝑛𝑗=1
(𝑓𝑖∗ −𝑓𝑖− ) Step 6: Prioritize the failure modes
(16)
The failure modes are prioritized based on
𝑤𝑗𝑜 (𝑓𝑖∗−𝑓𝑖𝑗 )
𝑅𝑖 = max ( ) (17) 𝑆𝑖 , 𝑅𝑖 , and 𝑄𝑖 in ascending order.
𝑖 (𝑓𝑖∗ −𝑓𝑖− )

*where 𝑤𝑗𝑜 the objective weight of risk

criterion obtained from Shannon entropy.

84
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
Step 7: Examine the condition of the
compromise solution
In this step, the condition of the compromise
solution is examined. Considering the minimum
value of 𝑄𝑖 , the failure mode 𝐹𝑀(1) is the first
ranked if the following two conditions are
fulfilled.
Condition #1: Acceptable advantage:
𝑄(𝐹𝑀(2) − 𝐹𝑀(1) ) ≥ 𝐷𝑄 where 𝐹𝑀(2) is the
second-ranked list by 𝑄𝑖 and 𝐷𝑄 = 1⁄𝐽 − 1.
Condition #2: Acceptable stability in the
decision making. The failure mode 𝐹𝑀(1) must be
the first priority ranked by 𝑆𝑖 or/and 𝑅𝑖 . It
indicates that the compromise solution is stable in
decision-making process (when 𝜐 > 0.5 is
required, or “by experts’ consensus” 𝜐 ≈ 0.5, or
“by veto” 𝜐 < 0.5).
If one of the conditions cannot be met, then
a set of compromise solutions is assigned as
follows:
- Failure mode 𝐹𝑀(1) and 𝐹𝑀(𝑚) if only if
“Condition# 2” is not met, or
- Failure mode 𝐹𝑀(1). , 𝐹𝑀(2). ,… , 𝐹𝑀(𝑚).
If only if “Condition# 1” is not met; 𝐹𝑀(𝑚) is
defined by the relation 𝑄(𝐹𝑀(m) − 𝐹𝑀(1) ) ≥
𝐷𝑄 for maximum 𝑀.
PROPOSED FRAMEWORK FOR
OPERATIONAL SUPPLY CHAIN
RISKS ASSESSMENT
This study proposed eight phases framework
for operational supply chain risks assessment
based on fuzzy FMEA as: Phase I-Identify
potential operational supply chain risks based on
FMEA, Phase II-Define risk criteria and
measuring scale, Phase III-Compute the
subjective weights of risk criteria, Phase IV:
85
Construct aggregated fuzzy rating matrix of FMs,
Phase V- Defuzzify the aggregated fuzzy rating
matrix of FMs, Phase VI-Compute the objective
weights of risk criteria, Phase VII-Prioritize the
FMs, and Phase VIII-Perform a sensitivity
analysis. The schematic diagram of the proposed
framework is illustrated in figure 2.
APPLICATION OF THE PROPOSED
FRAMEWORK
CASE STUDY
The proposed framework is validated by
using one of the leading canned tuna
manufacturers in Thailand. This manufacturer is
regarded as the largest producer of ready-to-eat
canned tuna products in Southeast Asia with
annual sales exceeding US$ 3.2 billion in the year
2020. The company operates as an original brand
manufacturer (OBM). In the past few years, this
company encounters various risks causing an
interruption in the supply chain. Company
executive staffs need to proactively develop an
effective plan to mitigate those risks. To do this,
a panel of experts consists of six experts with
more than ten years of experience including plant
manager, logistics manager, quality assurance
manager, risk management manager,
procurement manager and academician.
Therefore, the proposed framework in this study
is used as a tool to assist a panel of experts to
identify, assess and prioritize the risks inherited in
supply chain operations.
RESULTS
Phase I: Identify potential operational supply
chain risks based on FMEA
The experts are invited to investigate the
potential failure modes (FMs) of operation supply
chain risks for case manufacturing. After several
rounds of discussions, eleven FMs are identified
as presented in Table 2.
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Phase I

Identify potential operational supply chain risks based on FMEA

Phase II

Define risk criteria and measuring scale

Phase III

Compute the subjective weights of risk criteria

Phase IV

Construct aggregated fuzzy rating matrix of FMs

Phase V

Defuzzify the aggregated fuzzy rating matrix of FMs

Phase VI

Compute the objective weights of risk criteria

Phase VII

Prioritize the FMs

Phase VIII

Perform a sensitivity analysis

Fig. 2. The proposed framework

86
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
Code Failure Mode of operational supply
chain risks
FM1 Risk of volatility raw materials
supplied
FM2 Risk of relying on a few major
suppliers
FM3 Risk of product deterioration
FM4 Risk of variabilities in production
processes.
FM5 Risk of improper inventory
management
FM6 Risk of failing to comply with
industrial standard
FM7 Risk of inefficient traceability system
across supply chain processes.
FM8 Risk of a shortage of skilled workers
FM9 Risk of products damage and
contamination during transportation.
FM10 Risk of technological innovation
change.
FM11 Risk of failure in information
technology (IT) system
Phase II: Define risk criteria and measuring
scale.
Through brainstorming session, the experts
define the four risks of FMs as “severity” (S),
“probability of occurrence” (O), “probability of
detection” (D) and “degree of difficulty to
eliminate risks” (E). Also, the measurement
scales in linguistic terms for subjective important
weights of risk criteria, and for assessment of FMs
are provided in Table 3 and Table 4, respectively.
Table 2. The identification of potential FMs for case manufacturing
Effect
Manufacturers face a shortage of raw materials and/or raw materials that do
not meet the requirements. This results in excess lead times and
unacceptable quality levels of the raw materials supplied.
Manufacturers have low bargaining power with suppliers as their
dependence on a few major suppliers results in high raw material costs.
Manufacturers face deterioration and spoilage of tuna products caused by
the use of improper temperatures. Disease and contamination in transport
activities This results in higher cost of quality and loss of business
reputation.
Manufacturers face variability in the production process, resulting in higher
production costs, loss of productivity and non-conforming finished
products.
Manufacturers face higher inventory costs, inventory shortage and obsolete
inventory due to keeping too low or too high inventory.
Manufactures face revocation of required industry standard certificates
such as Good Manufacturing Practice: GMP, Hazard Analysis and Critical
Control Point: HACCP resulting in production halts and damage to
business reputation.
Manufacturers do not meet the security requirements for record storage and
counterfeit detection. This results in ongoing problems that are difficult for
manufactures to deal with in the event of a nonconforming product recall.
Manufacturers face a shortage of skilled workers. This can lead to a
competitive disadvantage in the seafood market.
Manufacturers face damage and product contamination during
transportation. This results in higher reverse logistics costs for
nonconforming products and customer complaints.
Manufacturers cannot keep up with the rapid changes of technological
innovations. This could lead to a competitive disadvantage in the seafood
market.
Manufacturers face disruptions in their businesses including sales,
production and cash flow in the supply chain due to IT system failure.
Phase III: Compute the subjective weights of risk
criteria
The experts evaluate the subjective weights
of risk criteria (S, O, D, E) by using linguistics
terms in Table 3 and the results are shown in
Table 5. The linguistic terms are converted to
their corresponding fuzzy numbers as shown in
Table 6. The subjective weights of risk criteria
(𝑤𝑗𝑠 ) are computed by aggregating the fuzzy
numbers by equation (2) and the results are shown
in Table 7.
87
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Table 3. The measurement scales for subjective important weights of risk criteria

Linguistic terms Abbreviation Fuzzy number

Very low VL (0.0,0.0,0.1,0.2)
Low L (0.1,0.2,0.2,0.3)
Medium low ML (0.2,0.3,0.4,0.5)
Medium M (0.4,0.5,0.5,0.6)
Medium high MH (0.5,0.6,0.7,0.8)
High H (0.7,0.8,0.8,0.9)
Very high VH (0.8,0.9,1.0,1.0)

Table 4. The measurement scales for assessment FMs

Severity Occurrence Detection Degree of difficulty to Trapezoidal

(S) (O) (D) eliminate risks fuzzy numbers
(E) (TFN)
No (N) Almost Never (AN) Almost Certain (AC) Almost no difficulty (N) (0,0,1,2)
Very Slight (VS) Remote (RS) Very High (VH) Remote (R) (0,1,2,3)
Slight (S) Very Slight (VS) High (H) Low (L) (1,2,3,4)
Minor (M) Slight (S) Moderately High (MH) Relative Low (RL) (2,3,4,5)
Moderate (MO) Low (L) Medium (M) Moderate (M) (3,4,5,6)
Significant (SI) Medium (M) Low (L) Moderate High (MH) (4,5,6,7)
Major (M) Moderate High (MH) Slight (S) High (H) (5,6,7,8)
Extreme (E) High (H) Very Slight (VS) Very High (VH) (6,7,8,9)
Serious (SE) Very High (VH) Remote (R) Extremely High (EH) (7,8,9,10)
Hazardous (H) Almost Certain (AC) Almost Impossible (AI) Almost Impossible (AI) (8,9,10,10)

Table 5: The evaluation of subjective weights in linguistic terms

Experts S O D E

𝐸1 VH MH M H

𝐸2 VH H ML MH

𝐸3 H MH M H

𝐸4 VH MH M H

𝐸5 VH MH ML MH

𝐸6 H MH M H

Table 6. The evaluation of subjective weights in fuzzy numbers

Experts Risk criteria

S O D E
𝐸1 0.8 0.9 1.0 1.0 0.5 0.6 0.7 0.8 0.4 0.5 0.5 0.6 0.7 0.8 0.8 0.9
𝐸2 0.8 0.9 1.0 1.0 0.7 0.8 0.8 0.9 0.2 0.3 0.4 0.5 0.5 0.6 0.7 0.8
𝐸3 0.7 0.8 0.8 0.9 0.5 0.6 0.7 0.8 0.4 0.5 0.5 0.6 0.7 0.8 0.8 0.9
𝐸4 0.8 0.9 1.0 1.0 0.5 0.6 0.7 0.8 0.4 0.5 0.5 0.6 0.7 0.8 0.8 0.9
𝐸5 0.8 0.9 1.0 1.0 0.5 0.6 0.7 0.8 0.2 0.3 0.4 0.5 0.5 0.6 0.7 0.8
𝐸6 0.7 0.8 0.8 0.9 0.5 0.6 0.7 0.8 0.4 0.5 0.5 0.6 0.7 0.8 0.8 0.9

88
 Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Table 7. The subjective weights of risk criteria (𝑤𝑗𝑠 )

Subjective S O D E
weights
𝑤𝑗𝑠 0.700 0.867 0.933 1.000 0.500 0.633 0.717 0.900 0.200 0.433 0.467 0.600 0.500 0.733 0.767 0.900

Phase IV: Construct aggregated fuzzy rating Phase V: Defuzzify the aggregated fuzzy rating
matrix of FMs matrix of FMs into crisp numbers

The experts employ the linguistic terms in
Table 4 to evaluate eleven FMs with respect to
risk criteria and the results are shown in Table 8.
The elements in Table 8 are then converted into
corresponding fuzzy numbers. Using equations
(3)-(4), the aggregated fuzzy rating matrix of FMs
is constructed as shown in Table 9.
According to the aggregated fuzzy rating
matrix Table 9, all elements are defuzzified into
crisp numbers by using equation (5). The
subjective weights can be obtained from Phase
III. Table 10 shows the crisp numbers of FMs
rating matrix.
Table 8. The evaluation of eleven FMs in linguistic terms

No. S O D E
FM1 SE,SE,SE,SE,SE,E H,VH,H,VH,H,VH M,MH,MH,MH,M,MH H,MH,MH,M,M,MH
FM2 E,MA,MA,SI,SI,MA M,M,MH,H,H,MH H,H,H,VH,VH,H MH,MH,H,H,MH,VH
FM3 SE,SE,SE,SE,SE,E H,VH,H,VH,H,VH H,H,H,MH,H,MH VH,EH,EH,VH,VH,EH
FM4 SI,MA,MA,SI,MA,MA H,VH,H,VH,H,VH H,MH,H,MH,MH,H H,MH,MH,M,MH,H
FM5 SI,MA,MA,MA,MA,SI M,M,L,L,M,MH H,H,MH,H,MH,MH H,H,MH,MH,M,M
FM6 SE,SE,SE,SE,SE,E M,L,L,L,M,L H,H,H,VH,H,VH H,H,MH,H,VH,VH
FM7 SI,SI,MA,MA,SI,MA MH,MH,M,H,MH,M H,M,MH,MH,M,MH MH,MH,H,H,MH,M
FM8 E,E,SE,MA,MA,E H,H,VH,MH,H,MH H,VH,H,VH,H,VH H,H,H,VH,MH,MH,H
FM9 E,SE,SE,E,E,SE M,L,L,L,L,M L,L,S,VS,S,M H,VH,VH,MH,H,MH
FM10 MA,MA,E,MA,MA,MA M,H,MH,MH,MH,MH H,H,H,VH,M,VH H,MH,MH,H,MH,MH
FM11 E,E,MA,MA,E,MA L,L,S,M,M,M H,H,VH,MH,V,M MH,MH,M,M,MH,H

Table 9. The aggregated evaluation decision matrix of FMs

S O D E
𝑥1 𝑥2 𝑥3 𝑥4 𝑥1 𝑥2 𝑥3 𝑥4 𝑥1 𝑥2 𝑥3 𝑥4 𝑥1 𝑥2 𝑥3 𝑥4
FM1 6.000 7.833 8.833 10.000 6.000 7.500 8.500 10.000 6.000 7.500 8.500 10.000 5.000 6.167 7.167 9.000
FM2 4.000 5.833 6.833 9.000 4.000 6.000 7.000 9.000 4.000 6.000 7.000 9.000 4.000 5.667 6.667 9.000
FM3 6.000 7.833 8.833 10.000 6.000 7.500 8.500 10.000 6.000 7.500 8.500 10.000 5.000 6.167 7.167 9.000
FM4 4.000 5.667 6.667 8.000 6.000 7.500 8.500 10.000 6.000 7.500 8.500 10.000 5.000 6.167 7.167 9.000
FM5 4.000 5.667 6.667 8.000 3.000 4.833 5.833 8.000 3.000 4.833 5.833 8.000 3.000 5.000 6.000 8.000
FM6 6.000 7.833 8.833 10.000 3.000 4.333 5.333 7.000 3.000 4.333 5.333 7.000 4.000 6.167 7.167 9.000
FM7 4.000 5.500 6.500 8.000 4.000 5.500 6.500 8.000 4.000 5.500 6.500 8.000 3.000 5.167 6.167 8.000
FM8 5.000 6.833 7.833 10.000 5.000 7.167 8.167 10.000 6.000 7.500 8.500 10.000 5.000 6.167 7.167 9.000
FM9 6.000 7.500 8.500 10.000 3.000 4.333 5.333 7.000 4.000 6.000 7.000 9.000 4.000 5.667 6.667 9.000
FM10 5.000 6.167 7.167 9.000 4.000 6.000 7.000 9.000 6.000 7.500 8.500 10.000 5.000 6.167 7.167 9.000
FM11 6.000 7.833 8.833 10.000 6.000 7.500 8.500 10.000 6.000 7.500 8.500 10.000 5.000 6.167 7.167 9.000

89
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Table 10. The crisp numbers of FMs rating matrix

Failure mode S O D E
FM1 8.133 8.000 5.438 6.867
FM2 6.435 6.500 2.067 6.370
FM3 8.133 8.000 5.933 6.867
FM4 6.067 8.000 5.933 6.867
FM5 6.067 5.435 3.000 5.500
FM6 8.133 4.933 2.067 6.565
FM7 6.000 6.000 3.565 5.565
FM8 7.435 7.565 2.000 6.500
FM9 8.000 4.933 6.000 6.500
FM10 6.867 6.500 2.810 5.933
FM11 7.000 4.630 2.937 5.435

performance rating (𝑓𝑖𝑗 ) of each FM with respect

Phase VI: Compute the objective weights of risk
criteria
In this study, Shannon entropy method is
employed to compute the objective weights of
risk criteria. Based on Table 10, the elements in
FMs rating matrix are normalized using equation
(6). The entropy measuring values (𝑒𝑗 ) the
divergence values (𝑑𝑗 ) of risk criteria are
computed using equation (7) and equation (8),
respectively. The objective weights (𝑤𝑗 ) of risk
criteria can be obtained by using equation (9).
Table 11 shows the results of objective weights
computation. It can be seen that the objective
weights of risk criteria be S (0.253) = E (0.253)>
O (0.251)>D (0.242).
Phase VII: Prioritize the FMs
In this study, fuzzy VIKOR is applied to
prioritize the FMs. Based on aggregated fuzzy
rating matrix in Table 9, the elements are
normalized using equation (10) for cost criterion
(S, O, E) and equation (11) for benefit criterion
(D). Table 12 shows the normalized aggregated
fuzzy rating matrix. Then, the overall
to risk criteria is calculated using equation (12),
as shown in Table 13. Based on Table 13, the best
ideal value (𝑓𝑗∗) and the worst ideal value (𝑓𝑗− ) of
FMs are determined by using equation (13) and
equation (14), respectively. The values of utility
measure 𝑆𝑖 , and regret measure 𝑅𝑖 for each FM
are obtained using equations (15)-(17),
respectively. Based on the objective weights
obtained from Phase VI, the 𝑄𝑖 value for each FM
is computed by using equations (18)-(22). In this
study, the 𝜐 value is defined as 0.5. Table 1 shows
the values of 𝑆𝑖 , 𝑅𝑖 and 𝑄𝑖 . Two conditions of
compromise solution are examined and the results
of both conditions are satisfied as shown in Table
14. Based on the results from Table 13 and Table
14, FMs are ranking, according to their 𝑄𝑖 values
in ascending order (the smaller value is the higher
the ranking). The results show that 𝐹𝑀3 >
𝐹𝑀1 > 𝐹𝑀4 > 𝐹𝑀9 > 𝐹𝑀8 > 𝐹𝑀10 >
𝐹𝑀6 > 𝐹𝑀2 > 𝐹𝑀7 > 𝐹𝑀11 > 𝐹𝑀5.
Therefore, FM3 is the most critical failure mode
and the tuna industry should give the first priority
to proactively manage risks.
Table 11. The object weights of risk criteria

S O D E
𝑒𝑗 5.733 5.703 5.538 5.741
𝑑𝑗 -4.733 -4.703 -4.538 -4.741
𝑤𝑗 0.253 0.251 0.242 0.253

90
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Table 12. The normalized aggregated fuzzy rating matrix

S O D E
𝑥1 𝑥2 𝑥3 𝑥4 𝑥1 𝑥2 𝑥3 𝑥4 𝑥1 𝑥2 𝑥3 𝑥4 𝑥1 𝑥2 𝑥3 𝑥4
1.500 1.958 2.208 2.500 3.000 3.750 4.250 5.000 0.222 0.537 0.648 1.000 1.667 2.056 2.389 3.000
FM1
1.000 1.458 1.708 2.250 2.000 3.000 3.500 4.500 0.000 0.185 0.296 0.444 1.333 1.889 2.222 3.000
FM2
1.500 1.958 2.208 2.500 3.000 3.750 4.250 5.000 0.444 0.593 0.704 0.889 1.667 2.056 2.389 3.000
FM3
1.000 1.417 1.667 2.000 3.000 3.750 4.250 5.000 0.444 0.593 0.704 0.889 1.667 2.056 2.389 3.000
FM4
1.000 1.417 1.667 2.000 1.500 2.417 2.917 4.000 0.111 0.278 0.389 0.556 1.000 1.667 2.000 2.667
FM5
1.500 1.958 2.208 2.500 1.500 2.167 2.667 3.500 0.000 0.185 0.296 0.444 1.333 2.056 2.389 3.000
FM6
1.000 1.375 1.625 2.000 2.000 2.750 3.250 4.000 0.111 0.352 0.463 0.667 1.000 1.722 2.056 2.667
FM7
1.250 1.708 1.958 2.500 2.500 3.583 4.083 5.000 0.000 0.167 0.278 0.444 1.333 2.000 2.333 3.000
FM8
1.500 1.875 2.125 2.500 1.500 2.167 2.667 3.500 0.333 0.611 0.722 1.000 1.333 2.000 2.333 3.000
FM9
1.250 1.542 1.792 2.250 2.000 3.000 3.500 4.500 0.000 0.222 0.333 0.667 1.333 1.778 2.111 2.667
FM10
1.250 1.625 1.875 2.250 1.000 2.167 2.667 3.500 0.000 0.259 0.370 0.667 1.000 1.611 1.944 2.667
FM11

Table 13. The overall performance rating (𝑓𝑖𝑗 ) of each FM

Failure mode S O D E 𝑆𝑖 𝑅𝑖 𝑄𝑖 Prioritization

𝑆𝑖 𝑅𝑖 𝑄𝑖
FM1 1.817 2.812 0.338 1.729 0.060 0.060 0.153 2 2 2
FM2 1.457 2.316 0.124 1.631 0.626 0.240 0.829 7 4 8
FM3 1.817 2.812 0.410 1.729 0.000 0.000 0.000 1 1 1
FM4 1.367 2.812 0.410 1.729 0.245 0.245 0.622 3 6 3
FM5 1.367 1.916 0.184 1.423 0.880 0.245 0.983 11 6 11
FM6 1.817 1.721 0.124 1.678 0.530 0.250 0.794 6 7 7
FM7 1.352 2.115 0.217 1.438 0.802 0.253 0.955 9 8 9
FM8 1.674 2.718 0.121 1.662 0.394 0.242 0.703 5 5 5
FM9 1.787 1.721 0.385 1.662 0.340 0.250 0.686 4 7 4
FM10 1.539 2.316 0.172 1.506 0.639 0.199 0.757 8 3 6
FM11 1.569 1.713 0.178 1.407 0.833 0.253 0.973 10 8 10

Table 14. The condition of compromise solution

Condition Check Result

Condition I Acceptable advantage 𝑄 (𝐹𝑀1) − 𝑄(𝐹𝑀3) ≥ 1⁄(𝑚 − 1)
0.153 − 0.000 ≥ 1⁄10 Satisfied
0.153 ≥ 0.1
Condition II Acceptable stability in FM3 is the best ranking indicated by 𝑄𝑖 , 𝑆𝑖 and 𝑅𝑖 Satisfied
decision making

different eleven scenarios are analyzed by varying

Phase VIII: Perform a sensitivity analysis
To evaluate the robustness of the proposed
FMEA model, the sensitivity analysis of the
decision is conducted by altering the different
values of weights for the strategy of maximum
group utility (𝜈). The objective is to examine
whether the FMs changes (according 𝑄𝑖 values)
when the 𝜈 value changes. In this study, a total of
𝜈 values to 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8,
0.9, and 1.0. The results of sensitivity analysis are
shown in Table 15 and figure 3. It can be seen that
FM3 is the most critical failure mode followed by
FM2 in all scenarios. Moreover, the overall
ranking of FMs is remained unchanged when ν =
0.4, 0.5 and 0.6, but it is sensitive at other 𝜈
values. It implies that the selection of 𝜈 values is
necessary to balance between utility measure (𝑆𝑖 )

91
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

and regret measure (𝑅𝑖 ) plays an important role should apply 𝜈 = 0.5, which means the failure
for implementing the proposed model. Thus, the modes are evaluated in a consensus way.
implementation of the proposed framework
Table 15. The results of sensitivity analysis

𝑄𝑖
Scenario 1 2 3 4 5 6 7 8 9 10 11
𝜐 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
FM1 0.238 0.221 0.204 0.187 0.170 0.153 0.136 0.119 0.102 0.086 0.069
FM2 0.946 0.923 0.899 0.876 0.852 0.829 0.806 0.782 0.759 0.735 0.712
FM3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
FM4 0.966 0.897 0.828 0.760 0.691 0.622 0.553 0.484 0.416 0.347 0.278
FM5 0.966 0.969 0.973 0.976 0.980 0.983 0.986 0.990 0.993 0.997 1.000
FM6 0.985 0.947 0.909 0.870 0.832 0.794 0.755 0.717 0.679 0.640 0.602
FM7 0.998 0.990 0.981 0.972 0.964 0.955 0.946 0.937 0.929 0.920 0.911
FM8 0.957 0.906 0.855 0.804 0.753 0.703 0.652 0.601 0.550 0.499 0.448
FM9 0.985 0.925 0.865 0.806 0.746 0.686 0.626 0.566 0.506 0.446 0.386
FM10 0.787 0.781 0.775 0.769 0.763 0.757 0.751 0.745 0.739 0.733 0.726
FM11 1.000 0.995 0.989 0.984 0.979 0.973 0.968 0.963 0.957 0.952 0.947

Table 16. The priority of FMs in each scenario

Priority of FMs
Scenario 1 2 3 4 5 6 7 8 9 10 11
𝜐 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
FM1 2 2 2 2 2 2 2 2 2 2 2
FM2 4 6 7 8 8 8 8 8 8 8 7
FM3 1 1 1 1 1 1 1 1 1 1 1
FM4 6 4 4 3 3 3 3 3 3 3 3
FM5 6 9 9 10 11 11 11 11 11 11 11
FM6 8 8 8 7 7 7 7 6 6 6 6
FM7 10 10 10 9 9 9 9 9 9 9 9
FM8 5 5 5 5 5 5 5 5 5 5 5
FM9 8 7 6 6 4 4 4 4 4 4 4
FM10 3 3 3 4 6 6 6 7 7 7 8
FM11 11 11 11 11 10 10 10 10 10 10 10

Fig. 3. Sensitivity analysis of eleven scenarios

92
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
CONCLUSIONS AND FUTURE
RESEARCH
In this study, a new FMEA based on MCDM
approach is proposed to assess operational supply
chain risks. The proposed framework can mitigate
the disadvantages of conventional FMEAs in a
number of ways. This study distinguishes from
previous researches in many aspects. First, the
fuzzy set theory under trapezoidal fuzzy number
is used to deal with the uncertain and imprecise
information in decision-making processes.
Second, the eleven failure modes of operational
supply chain risks are identified by a panel of
experts. Third, the new risk criteria namely
“degree of difficulty to eliminate risks” includes
risk assessment. Forth, the important weights of
risk criteria are determined by combining
subjective weights and objective weights. The
subjective weights are obtained by opinion of
experts, while the objective weights are derived
obtained by Shannon entropy method. Next,
fuzzy VIKOR is employed to prioritize failure
modes instead of a risk priority number (RPN).
Finally, a sensitive is performed and the results
indicate that the proposed framework provides the
stability and robustness for failure modes ranking.
A validation of the framework presented here uses
the canned tuna industry in Thailand as a case
study. The findings from this study indicates that
the most three critical FMs are “risk of product
deterioration” (FM3) followed by “risk of
volatility raw materials supplied” (FM1) and “risk
of variabilities in production processes” (FM4),
respectively. The outcomes of this study enable
tuna industry practitioners to proactively assess
the operational supply chain risks. Moreover, the
proposed framework can be applied to other
seafood industries in the same procedure. Further
research may extend from this framework by
investigating the interaction between risk criteria
using DEMATEL approach. Apart from that, the
other MCDM methods such as CRITIC can be
utilized to determine the objective weights. Also
future research should include the sustainability
dimensions of the supply chain in identifying
failure modes.
93
ACKNOWLEDGMENTS AND
FUNDING SOURCE DECLARATION
This research received no specific grant
from any funding agency in the public,
commercial, or not-for-profit sectors.
REFERENCES
Butdee S., Phuangsalee P., 2019. Uncertain risk
assessment modelling for bus body
manufacturing supply chain using AHP and
fuzzy AHP, Procedia Manufacturing, 30, 663–
670.
https://doi.org/10.1016/j.promfg.2019.02.094
da Silva C., Barbosa-P𝑜́ voa A.P., Carvalho A.,
2020. Environmental monetization and risk
assessment in supply chain design and
planning, Journal of Cleaner Production, 270,
121552.
https://doi.org/10.1016/j.jclepro.2020.121552
Fan H., Li G., Sun H., Cheng T.C.E., 2016. An
information processing perspective on supply
chain risk management: antecedents,
mechanism, and consequences, International
Journal of Production Economics, 185, 63-75.
http://doi.org/10.1016/j.ijpe.2016.11.015
Fattahi R., Khalilzadeh M., 2018. Risk evaluation
using a novel hybrid method based on FMEA
extended MULTIMOORA and AHP methods
under fuzzy environment, Safety Science, 102,
290-300.
http://doi.org/10.1016/j.ssci.2017.10.018
Heckmann I., Comes T., Nickel S., 2015. A
Critical Review on Supply Chain Risk -
Definition, Measure and Modeling, Omega,
52, 119-132.
https://doi.org/10.1016/j.omega.2014.10.004
Junaid M., Xue Y., Syed M.W., Li J.Z., Ziaullah
M., 2020. A neutrosophic AHP and TOPSIS
framework for supply chain risk assessment in
automotive industry of Pakistan,
Sustainability, 12(1), 154.
https://doi.org/10.3390/SU12010154
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
Karatop B., Taşkan B., Adar E., Kubat C., 2021.
Decision analysis related to the renewable
energy investments in Turkey based on a
Fuzzy AHP-EDAS-Fuzzy FMEA approach,
Computers & Industrial Engineering, 151,
106958.
http://doi.org/10.1016/j.cie.2020.106958
Lee H.C., Chang C. T., 2018. Comparative
analysis of MCDM methods for ranking
renewable energy sources in Taiwan,
Renewable and Sustainable Energy Reviews,
92, 883-896.
http://doi.org/10.1016/j.rser.2018.05.007
Liu B., Hu Y., Deng Y., 2018. New failure mode
and effects analysis based on D numbers
downscaling method, International Journal of
Computers, Communications & Control,
13(2), 205–220.
http://doi.org/10.15837/ijccc.2018.2.2990
Mete S. 2019. Assessing occupational risks in
pipeline construction using FMEA-based
AHP-MOORA integrated approach under
Pythagorean fuzzy environment, Human and
Ecological Risk Assessment: An International
Journal, 25(7), 1645-1660.
http://doi.org/10.1080/10807039.2018.15461
15
Mohamed E.A., Youssef, M.M., 2017. Analysis
of risk factors and events linked to the supply
chain: case of automotive sector in Morocco,
Journal of Logistics Management, 6(2), 41-51.
https://doi.org/10.5923/j.logistics.20170602.0
2
Moktadir M.A., Ali S.M., Mangla S.K., Sharmy
T.A., Luthra S., Mishra N., Garza-Reyes J.A.,
2018. Decision modeling of risks in
pharmaceutical supply chains, Industrial
Management & Data Systems, 118(7), 1388-
1412.https://doi.org/10.1108/IMDS-10-2017-
0465
94
Nabizadeh M., Khalilzadeh M., 2021. Developing
a fuzzy goal programming model for health
safety and environment risks based on hybrid
fuzzy FMEA-VIKOR method, Journal of
Engineering, Design and Technology, 19(2),
317-338. http://doi.org/10.1108/JEDT-09-
2019-0245
Nakandala D., Lau H., Zhao L., 2017.
Development of a hybrid fresh food supply
chain risk assessment model, International
Journal of Production Research, 55(15), 1-16.
https://doi.org/10.1080/00207543.2016.12674
13
Panchal D., Mangla S.K., Tyagi M., Ram M.,
2018. Risk analysis for clean and sustainable
production in a urea fertilizer industry,
International Journal of Quality and
Reliability Management, 35(7), 1459-1476.
https://doi.org/10.1108/IJQRM-03-2017-0038
Pourmadadkar M., Beheshtinia M.A., Ghods K.,
2020. An integrated approach for healthcare
services risk assessment and quality
enhancement, International Journal of Quality
& Reliability Management, 37(9/10), 1183-
1208. http://doi.org/10.1108/IJQRM-11-
2018-0314
Rathore R., Thakkar J.J., Jha J.K., 2021.
Evaluation of risks in foodgrains supply chain
using failure mode effect analysis and fuzzy
VIKOR, International Journal of Quality &
Reliability Management, 38(2), 551-580.
http://doi.org/10.1108/IJQRM-02-2019-0070
Sagnak M., Kazancoglu Y., Ozen Y.D.O., Garza-
Reyes J.A., 2020. Decision-making for risk
evaluation: integration of prospect theory with
failure modes and effects analysis (FMEA),
International Journal of Quality & Reliability
Management, 37(6/7), 939-956.
http://doi.org/10.1108/IJQRM-01-2020-0013
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645
Scheu M.N., Tremps L., Smolka U., Kolios A.,
Brennan F., 2019. A systematic failure mode
effects and criticality analysis for offshore
wind turbine systems towards integrated
condition based maintenance strategies, Ocean
Engineering, 176, 118-
133.https://doi.org/10.1016/j.oceaneng.2019.
02.048
Shaker F., Shahin A., Jahanyan S., 2019.
Developing a two-phase QFD for improving
FMEA: an integrative approach, International
Journal of Quality and Reliability
Management, 36(8), 1454-
1474.https://doi.org/10.1108/IJQRM-07-
2018-0195
Shan H., Li, Y., Shi J., 2020. Influence of Supply
Chain Collaborative Innovation on
Sustainable Development of Supply Chain: A
Study on Chinese Enterprises, Sustainability.
12 https://doi.org/10.3390/su12072978
Shannon C.E. (2001). A mathematical theory of
communication. ACM SIGMOBILE Mobile
Computing and Communication Review, 5(1),
3-55.https://doi.org/10.1145/584091.584093
Shemshadi A., Shirazi H., Toreihi M., Tarokh
M.J., 2011. A fuzzy VIKOR method for
supplier selection based on entropy measure
for objective weighting, Expert Systems with
Application, 38, 12160-12167
Shen L., Li F., Li C., Wang Y., Qian X., Feng T.,
Wang C., 2020. Inventory Optimization of
Fresh Agricultural Products Supply Chain
Based on Agricultural Superdocking, Journal
of Advanced Transportation, 1–13.
https://doi.org/10.1155/2020/2724164
95
Wang W., Lie X., Chen X., Qin Y., 2019. Risk
assessment based on hybrid FMEA framework
by consider maker’s psychological behavior
character, Computers & Industrial
Engineering, 136, 516-527.
https://doi.org/10.1016/j.cie.2019.07.051
Wu Y., Jia W., Li L., Song Z., Xu C., Liu F., 2019.
Risk assessment of electric vehicle supply
chain based on fuzzy synthetic evaluation,
Energy, 182, 397–411.
https://doi.org/10.1016/j.energy.2019.06.007
Yan H., Gao C., elzarka H., Mostafa K., Tang W.,
2019. Risk assessment for construction of
urban rail transit projects, Safety Science, 118,
583-594.
http://doi.org/10.1016/j.ssci.2019.05.042
Yazdi M., 2019. Improving failure mode and
effect analysis (FMEA) with consideration of
uncertainty handling as an interactive
approach, International Journal on Interactive
Design and Manufacturing, 13(2), 441–
458.https://doi.org/10.1007/s12008-018-
0496-2
Yazdi M., Nedjati A., Zarei E., Abbassi R., 2020.
A reliable risk analysis approach using an
extension of best-worst method based on
democratic-autocratic decision-making style,
Journal of Cleaner Production, 256, 120418.
https://doi.org/10.1016/j.jclepro.2020.120418
Yener Y., Can G.F., 2021. A FMEA based novel
intuitionistic fuzzy approach proposal:
Intuitionistic fuzzy advance MCDM and
mathematical modeling integration, Expert
Systems With Applications, 183, 115413.
http://doi.org/10.1016/j.eswa.2021.115413
Zadeh L., 1965. Fuzzy Sets. Information and
Control. https://doi.org/10.1016/S0019-
9958(65)90241-X
Sumrit D., Srisawad S., 2022. Fuzzy failure mode and effect analysis model for operational supply chain risks
assessment: an application in canned tuna manufacturer in Thailand, LogForum 18 (1), 77-96.
http://doi.org/10.17270/J.LOG.2022.645

Detcharat Sumrit ORCID ID: https://orcid.org/0000-0003-3157-6697

The Cluster of Logistics and Rail Engineering,
Faculty of Engineering, Mahidol University,
Nakhon Pathom, Thailand
e-mail: dettoy999@gmail.com

Sirima Srisawad
The Cluster of Logistics and Rail Engineering,
Faculty of Engineering, Mahidol University,
Nakhon Pathom, Thailand
e-mail: pijiro69@gmail.com

96